Gene and genomes in Coronary artery

disease

Yangsoo Jang

Yonsei University College of Medicine

Cardiovascular Center

Division of Cardiology

허혈성 심장질환 환자들에서의 주요 위험요인 동반개수

None

One

Two

Three Four (0.9%)

Total patients=87 869 CHD=coronary heart disease

†smoking, hypertension, hypercholesterolaemia and diabetes mellitus

19.4%

43.0%

27.8%

8.9%

Khot UN et al. JAMA 2003; 290: 898–904

1/3 to 1/2 of CHD disease occurs in patients classified as low risk by conventional Risk factors

Genetic basis of CHD

Twin studies indicate that heritability of CVD

is 30-60% (N Engl J Med 1994;330:1041)

Heritability as high as 63% for premature MI

(Circulation 1980;61:503)

History of premature death of a biological

parent(<50 years) was associated with 4.5

fold increase in mortality for adopted

offsprings Risk not increased in adopted

children with history of premature CAD in

foster parents (N Engl J Med 1988;318:727)

Human Genome Structure

chr #

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

X

Y

Size (Mb)

250

243

198

191

181

171

159

146

141

136

135

134

115

107

103

90

81

78

59

63

48

51

155

59

Genes #

3511

2368

1926

1444

1633

2057

1882

1315

1534

1391

2168

1714

720

1532

1249

1326

1773

557

2066

891

450

855

1672

429

Total genomic size = 3,106 Mb Total number of genes = 36,464 genes

Homo sapiens Build 37.3 (Oct. 5, 2011)

Human Genome Variations

Genome

Variation

Base

Variation

Single Nucleotide

Polymorphism (SNP)

Structural

Variation

Insertion & Deletion(InDel)

Copy Number Gain

Translocation

Inversion

Copy Number Loss

Copy Number Variation

(CNV)

Normal CNV,

Structural Variation

SNP, Haplotype and TagSNPs

Linkage Analysis (Family)

Association Study (Population)

• Single gene • Mendelian inheritance • Rare, but high penetrance • ~300-400 STR markers

• Polygenic (also G X E) • Complex inheritance • Common • Multiple polymorphic SNP markers

Strategies for Disease Gene Identification

Approaches to determine susceptibility genes

Candidate gene approach

GWAS

Catalog and test all coding SNPs for function

→ DNA Sequencing / SNapShot or TaqMan assay

5’ 3’

Use dense map of SNPs and test for LD (use association

to find sites in entire sequence with function)

→ Affymetrix GeneChip / Illumina GeneChip

▪ Candidate genes approach

▪ Genome-Wide Association approach

5’ 3’

Analytic Tools of Association Study

• GWAS are used to identify common genetic factors (SNP, Ins/del) that influence

health and disease.

General GWAS strategy for common complex disease

Genome-Wide Association Study (GWAS)

Linkage disequilibrium (LD) block & Association analysis

p value(-log transformed)

OR for CAD = 4.44(1.43-13.8) OR= 0.199(0.06-0.59)

(A)

(B)

In 2007: Discovery of 9p21 and the watershed moment for cardiovascular genetics

(A)

(B)

Human chromosome 9p21.3 (rs1333049)

had the strongest association with CAD

in WTCCC and Germans

Samani NJ et al. NEJM 2007;357:443-453

Role of 9p21 in the pathogenesis of CAD

Patel RS et al. Heart 2011;97:1463-1473

Ardissino D et al. J Am Coll Cardiol 2011;58:426-434

Hazard ratio of rs1333040

1.19 per risk allele(P = 0.01)

Survival estimation

According to presence of C allele Hazard Ratio

According to presence of C allele

Dutta A et al. Circ Cardiovasc Genet 2011;4:542-548

Interaction of 9p21 with diabetes and CVD

Case Control study of 734 type 2 Diabetics(322 CAD, 412 non CAD)

Doria A et al. JAMA 2008;300(20)

Chr. SNP Gene

(nearby)

Risk

allele

OR P vs.

KOR/EUR

6p24.1 rs6903956 C6orf105 A 1.71 5.0X10-3 N

9p21.3 rs1333048 CDKN2A/2

B

G 1.29 4.0X10-3 Y

Identification of significant association between 2 SNPs and CAD in Chinese

The function of C6orf105 gene is unknown, but Wang et al. suggested that decreased

expression of C6orf105 gene may be a possible pathogenic cause of CAD.

rs1333049 is associated with risk for

mi, but not with post-MI prognosis in

Han Chinese

Peng WH et al. CCLM 2009;47:917-22

G

Table. Association between haploypes with CAD in Koreans

This study provided solid evidence that the 4 SNPs on 9p21 are

associated with CAD in a South Koreans Shon GQ et al. ATVB 2008;28

Schematic Overview of Identification of CAD Causative SNPs

by GWAS [GenRIC Study]

Results of a meta-analysis for SNPs identified from both the GWAS and the replication cohorts

GWAS - Korea Replication - Japan Combined analysis

SNP Chromosome Gene Func Allele N OR P Allel

e N OR P OR p het.(P)

Previous publications

rs4537545 1q21.3e IL6R i T 4735 0.8356 2.39E-05 T 5234 0.8936 0.04297 0.8659 4.74E-05 0.3376

rs7588415 2p24.1c APOB A 4778 0.7578 2.74E-05 A 5233 0.8232 0.02587 0.7914 2.47E-05 0.4498

rs1333049 9p21.3c C 4770 1.263 3.30E-08 C 5000 1.47 1.8E-14

New identified loci

rs1111782 9p21.2a TEK i A 4716 0.8101 7.32E-07 T 5234 0.952 0.3684 0.8816 3.46E-04 0.0198

rs12114277 8q22.3b UBR5 i A 4674 0.8082 8.43E-07 A 5233 1.016 0.764 0.912 8.70E-03 9.00E-04

rs219822 7q22.1a TRRAP i A 4783 1.229 9.64E-07 A 5232 1.068 0.2277 1.1422 1.35E-04 4.12E-02

rs12705702 7q31.1b T 4781 0.8263 4.17E-06 G 5232 1.039 0.4792 0.9314 4.06E-02 8.00E-04

rs1163072 10q24.33 T 4780 1.203 1.04E-05 G 5231 1.051 0.3571 1.1208 9.86E-04 0.0487

rs41391154 3p26.1a GRM7 i T 4775 0.7166 1.24E-05 T 5233 0.9898 0.9076 0.8487 5.05E-03 0.0055

rs886126 12q24.11d CUX2 i C 4756 0.8244 1.31E-05 C 5232 0.9654 5.45E-01 0.8958 2.92E-03 0.0309

rs10012505 4q34.1b GALNT17 i G 4662 0.7661 1.67E-05 C 5233 0.9422 0.4594 0.8547 2.35E-03 0.0416

rs2122149 4q13.1a A 4674 1.277 1.87E-05 A 5231 1.03 0.6516 1.14 2.96E-03 0.0138

rs9944810 18q21.31 ALPK2 cn C 4783 0.8326 2.08E-05 C 5234 0.9488 0.3364 0.8914 1.09E-03 0.0603

SNP1 12 MYL2 i C 4762 1.255 2.13E-05 G 5232 1.262 9.85E-05 1.2586 1.13E-08 0.9446

SNP2 13 FLT1 i C 4789 1.192 2.34E-05 G 5231 1.148 1.09E-02 1.1688 6.35E-06 0.5817

SNP3 4 IGFBP7 i G 4781 1.187 3.27E-05 C 5233 1.116 0.04156 1.1491 4.91E-05 0.3625

Hum Mol Genet [IF 8.058] Submitted

Association of rs1333049 on chromosome 9p21 with CAD in Japanese and Koreans

Chr.9p21 rs1333049 was the susceptibility locus for CAD

in Japanese and Koreans

17 Genetic Loci associated with CAD

Sivapalaratnam S et al. Curr Atheroscler Rep 2011;13:225-232

Hum Mol Genet 2011;20:R198

ANRIL important for expression of CDK activity

and vascular proliferation

Pasmant E et al. FASEB J 2011;25:444-448

Reilly M et al. Lancet 2011;377:383-392

Odds Ratio of CAD According to

Presence of ADAMTS7 variant

rs514659 at 9q34.2 at ABO locus and MI Reilly M et al. Lancet 2011;377:383-392

Odds Ratio of MI According to

Presence of rs514659 variant at ABO

3rd Validation study [CGC Cohort] (CAD : control = 502:648)

Confirmation of Korean specific SNPs

Cumulative association analysis

4th Validation study (CAD progress)

Identify significant SNPs

Clinical follow up study

#38

GWAS of CAD in CGC study

Conventional Factors

Biomarkers SNP (32)

Age BUN ADIPOQ rs1801133 rs2569190 rs1333049 rs4341

Sex Creatinine LP-PLA2 rs17465637 rs10946398 rs2230806 rs1502017

DM Glucose RAGE rs12713259 rs909253 rs4149263

HT Insulin IL-6 rs16944 rs2070600 rs13290387

Smoke hsCRP RANTES rs1143623 rs1051931 rs501120

BMI rs4848306 rs16874954 rs5015480

T-Chol rs4402960 rs1871388 rs5215

HDL rs2241766 rs1800796 rs11048979

TG rs1501299 rs13266634 rs8050136

LDL rs155948 rs564398 rs2107538

CAD group (503 명)

* Male: ≥55 years, Female: ≥60 years

* Multi-vessels

Control group (503 명)

* Case group과 1:1 matching

* Control 146명 추가 분석

Construction of CAD-related CGC Cohort

GWAS Repl. 1 Repl. 2 Repl. 3

Case / Control 230/290 1392/1355 502/648 2123/2690

Mean age(case) 48.3±4.72 54.5 ±7.79 49.3 ±5.36 51.6±7.52

Age criteria ≤55 ≤65 ≤55(m), 60(f) ＜55(m), 60(f)

Chr. Gene rs # Stage1 (500K) Stage2 (3K) Stage3 (32) Stage4 (15)

9p rs133**** 5.643E-05 3.080E-07 0.0109 3.30E-08

2p SPTBN1 rs127**** 3.009E-06 1.060E-05 0.0048 0.0034

5q rs15**** 3.125E-03 1.770E-05 0.0049 0.0455

12p ARNTL2 rs1104**** 8.994E-04 6.920E-06 0.0047 0.5961

19p CACNA1A rs150**** 4.839E-05 4.320E-05 0.0890 -

1p Vav3 rs1275**** 6.192E-03 1.906E-03 0.3889 -

1p rs599*** 0.0272 - 0.8887 -

1q MIA3 rs1746**** 0.0860 - 0.0227 0.0934

10q rs501*** 0.9211 - 0.0100 -

Results of CAD-GWAS in CGC

13

The most notable association with CAD was observed on chromosome 9p21.3.

The strongest signal was at rs1333049.

These results replicate signals first observed in Caucasians and subsequently

observed in a variety of Asians including Japanese, Korean and Chinese Han.

Cho E et al. Heart Asia 2010;104-108

Genetic Risk Factors

New Risk Prediction by Biomarkers

Genetic Analysis of CAD susceptible SNPs

- Chr. 2 (1 SNP) - Chr. 5 (1 SNP)

- Chr. 9 (1 SNP) - Chr. 12 (1 SNP)

- Chr. 19 (1 SNP) - LTA (1 SNP)

- AdipoQ (2 SNPs) - RAGE (1 SNP)

- Lp-PLA2 (2 SNPs) - IL-6 (1 SNP)

- RANTES (1 SNP) - PAPPA (1 SNP)

- IL-1β (3 SNPs) - CD14 (1 SNP)

- ABCA1 (1 SNP) - ACE (1 SNP)

- MTHFD1 (1 SNP) - Chr.2 (1 SNP)

- PSRC1 (1 SNP) - MIA3 (1 SNP)

- Chr.10 (1 SNP) – SMAD3 (1 SNP)

- FTO, CDKAL1, HHEX, CDKN2B,

IGF2BP2, SLC30A8, KCNJ11

- MTHFR

Interpretation of Risk Prediction

Probability (%)

Myocardial infarction ?

CAD ?

Individuals

Age, Sex, BMI, Smoking status,

HTN, DM, Glucose, Insulin, TG,

LDL-C, HDL-C, hsCRP, BUN,

Creatinine

Laboratory Test

Polymorphic

markers

Conventional Risk Factors

HOMA-IR

- RANTES

- RAGE

- IL-6

- Adiponectin

- Lp-PLA2

Contents for Personalized

Genetic Risk Predictive

System for CAD

개선된 CAD 발생 위험 예측 모형 구축

CAD 발생 위험 확률 =

TT TC TT

TC CC CG

TT

GRS 0.8236 {(-0.0189 ARNTL2 )-(0.0456 ARNTL2 )-(0.1160 CD14 )-

(0.1213 CD14 ) (0.0077 PAPPA )-(0.0387 PAPPA )-

(0.0275 ABCA1 )-(0.0428

TCABCA1 )-(0.0440 DM)+

(0.0143 HT)- (0.0245 Smoke)+(0.078 BMI)-(0.2625 T-chol)+

(0.0981 HDL)+(0.3092 LDL)+(0.1527 Tg)+(0.2174 Creatinine)

(0.4784 Glucose)-(0.0889 hsCRP)-(0.0007 Adiponectin)}

CAD 발생 위험 확률 예측을 위한 2-stage model 구축(1,006명)

exp( 4.0629 0.0622 )

1 exp( 4.0629 0.0622 )

GRS

GRS

CAD 발생 위험 확률 =

CAD 발생 위험 예측시스템의 프로그램 실행 화면

CAD 발생 위험 예측시스템 개발

Two-stage model을 이용한 CAD 발생 위험 확률 계산 화면

One-stage model을 이용한 CAD 발생 위험 확률 계산 화면

예측모형을 통해 추정된 CAD 발생 위험 확률 (예)

DM Smoke BMI … hsCRP ARNTL2 CD14 PAPPA ABCA1 Probability(%)

No Yes normal … normal TC TC CG TC 12.33

No Yes normal … normal TC TT CG TC 12.35

No Yes normal … normal TC TC CG TT 13.86

…

No Yes normal … normal TT CC CC CC 50.22

No Yes normal … normal CC CC GG CC 50.61

No Yes normal … normal CC CC CC CC 53.90

Yes No normal … normal TC TC CG TC 12.54

Yes No normal … normal TC TT CG TC 12.56

Yes No normal … normal TC TC CG TT 14.10

…

Yes No normal … normal TT CC CC CC 50.71

Yes No normal … normal CC CC GG CC 51.10

Yes No normal … normal CC CC CC CC 54.39

CAD 예측 실용 모형

SNPs

rs11***(ARNTL2), rs12***(SPTNB1), rs15***(Ch5), rs13***(Ch9)

Conventional risk factor

DM, hypertension, smoke, BMI, Total-cholesterol, Tg, HDL, Creatinine

민감도 80.4%, 특이도 84.5%,

정확도 81.5%, AUC 0.906

타당성 평가

(GenRIC + KCDC 자료) 민감도 100%, 특이도 73.5%, 정확도 86.5%

CAD 발생 위험 예측모형 개발 결과

Loci SNP OR P Gene Function

2p21 rs1687**** 1.212 3.01E-06 SPTBN1 Determination of cell

shape, arrangement of

transmembrane proteins

4q12 rs2124*** 1.187 3.27E-05 IGFBP7 Modulation of vascular

remodeling

5q rs1507*** 1.285 3.13E-03 intergenic Not known

9p21.3 rs1333*** 1.263 3.30E-08 nearby

CDKN2A/2B

Regulation of the cell cycle

12q23 rs3782*** 1.255 2.13E-05 MYL2 Regulation of myosin

ATPase activity in smooth

muscle

13q12 rs9508*** 1.192 2.34E-05 FLT1 Control of cell proliferation

and differentiation

Summary: Identified causative genetic variants for CAD in Korans

3rd Validation study [CGC Cohort] (CAD : control = 502:648)

Confirmation of Korean specific SNPs

Cumulative association analysis

4th Validation study (CAD progress)

Identify significant SNPs

Clinical follow up study

국내특허 출원 (No. 10-2008-0109058)

#38

신의료기술평가 (건강보험심사평가원)

Heart Asia (2(1) 104-108, 2010)

향후 심혈관질환 예측모형 콘텐츠 발굴 전략

[1] CAD GWAS (CGC) – 상위 20종 유전자 [2] CAD GWAS (GenRIC) – 상위 20종 유전자 CAD case-control study (CGC)

Candidate gene approach를 통해 유의한 연관성을 나타난

유전자 가운데 GWAS SNP chip에 포함되지 않은 유전자 20종

GenRIC study에서 발굴된

유전자 20종 선발

Target capture NGS analysis

SNP contents validation

Modest effect of common alleles

Only 1 study from the

ARIC study showed improvement

Of CV risk predictability

Feasibility of identifying genetic variants by risk allele frequency and strength of genetic effect

Manolio et al, Nature 2009

We have only touched the tip of the iceberg

Common Variants Rare Variants? Structural

variants?

Epigenetic

Phenomenon?

Environmental

Influence

Full genetic influence

on Hypertension

a) Sanger method (old), b) NGS (new) Chip-based capturing exome sequencing

Next Generation Sequencing (NGS)

Gap between hereditability and effect

More common variants that need to be

discovered

Gene-gene interaction, gene-environment

interaction

Effect of numerous rare variants(0.5-5% minor

allele frequency) with large effects NGS

Effect of structural variants in Hypertension

pathogenesis NGS

Effect of chromosomal structure

Epigenomics

Conclusion & Further Study

Identified several genetic loci that were strongly

associated with CAD in worldwide studies by GWAS

Further studies are needed to survey the associations

of the loci with other types of atherosclerotic disease

At a genetic level, studies should focus on fine mapping

of the associated regions using NGS

Targeted sequencing of CAD-associated multi-gene by NGS

is needed for further identification of causative rare variants

Yonsei Cardiovascular Hospital

Yonsei University College of Medicine

Loci SNP OR P Gene Validated

population

6p21.2 rs16874954

(V279F)

1.922 0.013 Lp-PLA2 Chinese

0.80 0.002 Korean

6p21.32 rs2070600

(G82S)

2.303 0.001 RAGE Chinese

0.749 0.028 Korean

12q24.12 rs11066001 1.63 5.0X10-11 BRAP Japanese

1.68 6.5x10-9 Korean

Validated SNPs identified in Asian candidate gene

studies for CAD